The overall objective of this project is to provide an advanced animal whole body/human brain positron tomograph for the medical science projects. This tomograph will make possible quantitative measurements of cardiovascular physiology in animal models by permitting high count rate dynamic studies with bolus injections of flow and metabolic tracers. It has the following advanced features: Lutetium orthosilicate (LSO) scintillators, which have 6 times greater light output and 8 times faster decay time than bismuth germanate (BGO). Photomultiplier tube readout to provide pulse height information (to reject scattered annihilation photons) and a timing resolution less than 1 ns (to reject random backgrounds). Small (1.5 - 2.5 mm) crystals, read out individually by low-noise silicon photodiodes. Depth of interaction measurement for less than 2 mm spatial resolution over the imaging volume. Small detector dimaeter of 35 cm for low noncollinearity error, high sensitivity, and reduced cost. Full 3D mode over a subject port of 30 cm and an axial field of 15 cm (no inter-plane septa), for high sensitivity and a maximum noise equivalent event rate of greater than 700 k/s at 1 muCi/m1. We expect that U.S. Department of Energy funding will support purchase of the components needed for the 270 detector modules, that CTI PET Systems will provide the labor to assemble and test them, and that this project will provide the following: Construction, testing, and evaluation of pre-production detection modules. Assembling 270 pre-tested detector modules in a gantry at LBL. Evaluating the completed high resolution tomograph. Developing attenuation measurement, background correction, and calibration procedures. This project will advance the technology further by developing time-of-flight capability and installing it in the high resolution LSO tomograph. Our goal is a timing resolution less than 500 ps, which corresponds to less than 7.5 cm. This improves the signal-to-noise ration in the reconstructed 3D images by further reducing the random coincidence background and by localizing the events in 3D to reduce error propagation in the reconstruction process.